WO2007033959A2

WO2007033959A2 - Use of p-selectin glycoprotein ligand 1 in modulation of hematogenous metastasis of lymphomas

Info

Publication number: WO2007033959A2
Application number: PCT/EP2006/066490
Authority: WO
Inventors: Patrick De Baetselier; Geert Raes; Gholamreza Hassanzadeh Ghassabeh
Original assignee: Vib Vzw; Vrije Universiteit Brussel
Priority date: 2005-09-19
Filing date: 2006-09-19
Publication date: 2007-03-29
Also published as: WO2007033959A3

Abstract

The present invention relates to the use of P-selectin ligand 1 for the modulation of hematogenous metastasis of lymphomas. More specifically, it relates to the use of P-selectin glycoprotein ligan-1 and of molecules that impair the binding of P-selectin ligand in preventing metastasis, preferably metastasis in liver and spleen.

Description

USE OF P-SELECTIN GLYCOPROTEIN LIGAND 1 IN MODULATION OF HEMATOGENOUS METASTASIS OF LYMPHOMAS

P-selectin glycoprotein ligand-1 (PSGL-1 ) is a mucin-like sialylated surface membrane- associated glycoprotein that was shown to bind to not only P-selectin, but also E- and L- selectin (Moore et al., 1992; Sako et al., 1993; Walcheck et al., 1996) and to be broadly expressed on almost all leukocytes, including cells of myeloid, lymphoid and dendritic lineage (Laszik et al., 1996). Yet, its ability to act as a functional ligand for the different selectins is dependent on various posttranslational modifications and differs among cell types and differentiation or activation stages (Moore et al., 1995; Vachino et al., 1995; Borges et al., 1997). This sialomucin selectin ligand is involved in the initial stages of leukocyte recruitment to sites of inflammation, by promoting among others tethering and rolling of leukocytes on activated endothelial cells and platelets (McEver and Cummings, 1997). Interactions of PSGL- 1 with E-selectin also participate in homing of hematopoietic progenitor cells to bone marrow, by contributing to progenitor rolling in the bone marrow microvasculature (Katayama et al, 2003). Moreover, besides acting as an adhesion molecule, PSGL-1 binds to chemokines, suggesting a role in regulating chemokine-mediated responses (Hicks et al., 2003; Hirata et al., 2004), and functions as a signal-transduction receptor, reportedly modulating responses of PSGL-1 -expressing T-cells (Chen et al., 2004; Atarashi et al. 2005) and neutrophils (Ba et al., 2005). Migration of leukocytes throughout the organism and subsequent infiltration of the local tissue bears resemblance to the migratory and disseminative behavior of malignant leukemias and lymphomas. Hence, these two processes may obviously share some common underlying mechanisms and molecular determinants. Moreover, expression of the normal migratory properties of leukocytes by their neoplastic counterparts may result in tumor variants with metastatic potential. We have studied this phenomenon before, using a tumor model based on the murine T-lymphoma BW5417. These cells are not invasive in two different in vitro invasion assays and, following intravenous (i.v.) inoculation, do not form detectable metastases in most recipients. Yet, invasive and metastatic variants of the BW5147 cell-line have been derived in various ways (the initial, parental cell-line will further be referred to as BW-O). For instance, in vitro or in vivo somatic cell fusion with lympho-reticular cells such as T-cells, B-cells and macrophages can yield hybridomas that have acquired in vitro invasive properties and form massive metastases in vivo in organs such as spleen or liver (De Baetselier et al., 1984 a, b). Alternatively, invasive and metastatic properties can be induced in BW-O by conditioning in the local tumor micro-environment upon subcutaneous (s.c.) or intrasplenic (i.s.) inoculation (De Baetselier et al. ,1988).

As documented before (Van Hecke et al., 1990), selected pairs of metastatic and non- metastatic BW-O-derived T-cell hybridomas were used to generate and screen monoclonal antibodies (MAbs) recognizing membrane markers expressed selectively on the invasive and metastatic T-cell hybridomas. Three such MAbs were obtained that recognize an identical membrane-associated sialoglycoprotein, termed the "metastatic T-cell hybridoma antigen" (MTH-Ag). Expression of this MTH-Ag correlated directly with the experimental metastatic capacity of BW-O-derived T-cell hybridomas and lymphomas, as well as BW5147-unrelated T- lymphomas (De Baetselier et al., 1988; Van Hecke et al, 1990). Yet, the molecular identity of MTH-Ag remained elusive and it was unclear whether MTH-Ag played a causative role in the metastatic process or whether its expression was merely associated with progression to a metastatic phenotype. Surprisingly we found that the MTH-Ag is PSGL-1 and that it is a critical determinant of the disseminative properties of the BW-O-derived T-cell hybridomas and lymphomas, particularly implied in liver colonization, and to a lesser extent in spleen colonization. A first aspect of the invention is the use of PSGL-1 to modulate metastasis of lymphoma, preferably a lymphoma characterized in by expression of PSGL-1 (previously MTH-Ag), which can be measured as described by Van Hecke et al. (1990). Preferably, said lymphoma is a non-Hodgkin's lymphoma. Preferably, said modulation of metastasis is a repression or reduction of the metastasis, even more preferably repression or reduction of liver or spleen metastasis, most preferably repression or reduction of liver metastasis. Said repression or reduction of the metastasis can be realized either by limiting the expression of PSGL-1 or by impairing or inhibiting the binding capacity of PSGL-1 . Limitation of expression can be obtained, as a non-limiting example, by inactivating the PSGL-1 encoding gene in the tumor cells, by inactivation of the promoter of the PSGL-1 encoding gene in the tumor cells or by expressing PSGL-1 RNAi in the tumor cells.

As a similar result on metastasis can be obtained by impairing the binding capacity of PSGL-1 , another aspect of the invention is the repression or reduction of metastasis of lymphoma, preferably liver or spleen metastasis, more preferably liver metastasis, by inhibiting the binding capacity of PSGL-1 . Methods to inhibit the binding capacity of PSGL-1 are known to the person skilled in the art and include, but are not limited to the use of anti-PSGL-1 antibodies, soluble PSGL-1 , soluble P-selectin and mocarhagin. Antibodies against P-selecting glycoprotein ligand, which inhibit the binding have been disclosed in US2003072755. Several authors described (recombinant) soluble PSGL-1 and its effect on PSGL-1 binding (Hayward and Lefer, 1999; Kyriakides et al., 2001 ; Wang et al., 2002). Soluble P-selectin has been disclosed by Croce et al. (1998) and it was shown that it kept its binding capacity to PSGL-1 ; therefore soluble P-selectin would act as a competitive inhibitor of the binding of PSGL-1 with cell bound P-selectin or another, cell bound receptor. Mocarhagin has been disclosed in WO9705244 and WO9846771 , where it was shown to cleave PSGL-1 and to disrupt PSGL-1 mediated cell adhesion. However, none of these PSGL-1 binding inhibitors has been described as repressors of metastasis in general of metastasis of lymphomas in particular.

BRIEF DESCRIPTION OF THE FIGURES

Figure Legends

Figure 1 . Comparative expression analysis of MTH-Ag and PSGL-1 on BW5147 variants. (1 A) BW5147 variants with an indication of their metastatic capacity (M) upon i.v. inoculation, expressed as the extent of colonization of spleen (Msp) and liver (MIi). -: no colonization; +/-: limited colonization (< 20% increase in organ weight); ++: heavy colonization (>100% increase in organ weight). (1 B) Expression of MTH-Ag, determined by indirect immunofluorescence and flowcytometric analysis, using anti-MTH-Ag antibodies as primary antibodies and FITC- conjugated goat anti-rat IgG as secondary antibodies. The dotted peak corresponds to the background profile of cells stained without primary antibodies. (1 C) Expression of PSGL-1 , determined by direct immunofluorescence and FACS analysis, using PE-labeled anti-PSGL-1 antibodies. The dotted peak corresponds to the background profile of cells stained with PE- labeled isotype-matched control antibodies. Figure 2. Detection of MTH-Ag and PSGL-1 on PSGL-1 transfectants of BW-N and BW-Sp3. The represented signals are the background-subtracted median fluorescence intensities, obtained after indirect (anti-MTH-Ag MAbs; black bars) or direct immunofluorescence (anti- PSGL-1 MAbs; white bars) and flowcytometric analysis. Cells stained without primary antibodies (MTH-Ag) or with isotype-matched control antibodies (PSGL-1 ) were used to determine background levels.

Figure 3. Expression of PSGL-1 on representative siRNA-transfected clones of BW-O-TUM with a reduction in MTH-Ag/PSGL-1 expression of less than 10% (BW-TR10), about 50% (BW- TR50) or more than 90% (BW-TR90) as compared to the parental BW-O-TUM. Expression was determined by direct immunofluorescence and flowcytometric analysis. The dotted peak corresponds to the background profile of cells stained without primary antibodies. Numbers between brackets indicate the background-subtracted median fluorescence intensities. Figure 4. Effect of PSGL-1 siRNA transfection on malignancy of BW-O-TUM upon i.v. inoculation, represented as survival of mice in function of time after inoculation (4A) and colonization of liver (4B) and spleen (4C). Survival curves were obtained using groups of 9 mice each. Organ weight was determined using separate groups of mice sacrificed at 17 days post inoculation, except for the data in the white bars, representing organ weight at the time of mortality. The dotted line represents the weight of organs from control non-tumor-bearing mice. (^*): Significantly reduced as compared to BW-O-TUM (p<0.0001 ).

EXAMPLES Materials and methods to the examples Mice and cell-lines

Specific pathogen-free female AKR/OlaHsd mice (Thy 1 .1 , H-2^k) were purchased from Charles River Laboratories. BW5147 (also referred to as BW-O) is a T-cell lymphoma of AKR origin (SaIk Institute, La JoIIa, CA). Different variants of BW-O, obtained by various in vitro and in vivo procedures, have already been described (De Baetselier et al., 1984b; De Baetseleir et al, 1988; Vanden Driessche et al., 1990).

Generation of constructs

The full-length PSGL-1 cDNA (including the signal peptide) was obtained by reverse transcription-PCR on total RNA from BW-19 cells using sense primer 5'- CGGGGTACCGTACCATGTCCCCAAGCTTC-3' and anti-sense primer 5'- GCTCTAGAGTGGAGCTAGCAAAGGTCTC-3' (the restriction sites are underlined). PCR products were cloned into the Kpnl and Xbal sites of the pcDNA3.1 (+)Neo plasmid (Invitrogen). The insert sequence was verified to confirm that the amplified cDNA matched the NCBI RefSeq sequence (NM_009151 ) of murine PSGL-1.

The short interfering RNA (siRNA) sense sequences and the siRNA hairpin constructs were designed using the siRNA Target Finder and the siRNA Construct Builder softwares (Genscript), respectively. The three siRNA hairpin constructs used were GGATCCCGLACIG/AGGILAG/ACICC/ACiαtαatatccαCAGTGGAGTCTAACCTCAGTATTTT TTCCAAAAGCTT (construct No. 1 ),

GGATCCCGILAG/ACICIGIGG/ACGCCGiαtαatatccαGACGGCGTCCACAGAGTCTAATTT TTTCCAAAAGCTT (construct No. 2) and GGATCCCGICGGIGGCIGCΛGΛCGIIGIAttαatatccαTACAACGTCTGCAGCCACCGATTT TTTCCAAAAGCTT (construct No. 3) (BamHI and Hindlll restriction sites are underlined, the anti-sense and sense sequences are, respectively, in italic and bold, the loop sequence is in lower case, and the 6 Ts following the sense sequence serve as termination signal). The siRNA hairpin constructs were chemically synthesized by Genscript, cloned into the BamHI and Hindlll sites of the pRNAU6.1/Neo plasmid (Genscript) and sequence verified.

Transfection

For stable transfection, 2 x 10⁷ cells were mixed with 20 μg of vector DNA and electroporated, using a Biorad Gene Pulser (300 V, 960 μF). For control transfections with empty vector, 2 μg of vector DNA was used. Two days later, cells were plated out in 96-well plates (Falcon) in medium supplemented with 5 mg/ml G418 (Geneticin; Gibco). Individual antibiotic-resistant colonies were expanded in medium without antibiotics and screened for surface expression of MTH-Ag/PSGL-1 .

Detection of surface antigen expression in flow cytometry

Cytofluorimetric analysis was performed as described previously (Van Hecke et al, 1990). For evaluation of MTH-Ag expression, cells were stained indirectly with an optimal dilution of a pool of the 3 anti-MTH-Ag MAbs as primary antibodies and FITC-conjugated goat anti-rat IgG (Sigma) as secondary antibody. Expression of PSGL-1 was determined by direct immunofluorescence, using PE-labeled anti-PSGL-1 antibodies (BD Pharmingen). Expression of LFA-1 and H-2D^k were assessed using as primary antibodies biotinylated anti-LFA-1 (BD Pharmingen) and anti-H-2D^k (15-5-5S; ATCC HB-24), followed by staining with FITC- conjugated streptavidin (BD Pharmingen) and FITC-conjugated goat anti-mouse IgG (ICN Flow), respectively. Expression levels were determined with a FACSVantage station (BD Biosciences) and data were analyzed with CellQuest software.

Evaluation of malignancy The experimental metastatic potential was assessed by injecting mice intravenously via the tail vain with 10⁶ cells and monitoring survival of the recipients. Moribund mice were killed, metastases were evaluated macroscopically and weight of kidneys, liver, spleen and lungs was determined. Organ weight was also determined for mice in additional groups, sacrificed at the time the mice inoculated with the parental cell-line started dying.

Statistical analysis

Significance of differences in survival curves was assessed via the log-rank test and all pair- wise comparisons (organ weights, expression levels in flow cytometry) were tested for statistical significance (p < 0.05) via the unpaired t test, using GraphPad Prism 3.0 software. For each experimental setup, results were confirmed in at least two independent experiments, each involving at least three mice per group.

Example 1 : The molecular properties and expression patterns of the MTH-Ag are similar to those of PSGL-1 The first attempts at molecular characterization of the MTH-Ag had revealed that it is a sialylated glycoprotein with an apparent molecular weight of about 95-100 kDa, as determined via Western blotting under reducing conditions (Van Hecke et al, 1990). Additional experiments have shown that, depending on the reducing conditions used, a minor band of about twice that size is often detected and that under non-reducing conditions this band of about 220 kDa is the main species detected (data not shown). This is similar to PSGL-1 , which was reported to be a sialomucin composed of two disulfide-linked subunits of -120 kDa, with relatively high resistance to complete reduction (Moore et al., 1992; Fuhlbrigge et al., 1997). Moreover, besides on metastatic T-cell hybridomas and lymphomas, the MTH-Ag was originally reported to be expressed on normal T-lymphocytes (Van Hecke et al, 1990). We have later extended this observation to IL-2-stimulated lymphokine activated killer cells, B-cell hybridomas, CD1 1 b- positive bone marrow cells, macrophage cell-lines and peritoneal macrophages, indicating a broad expression of MTH-Ag on hematopoietic cells. Again, this is similar to the reported expression patterns of PSGL-1 on immune cells (Laszik et al, 1996).

Due to these similarities in molecular properties and expression patterns, we were wondering if the MTH-Ag was PSGL-1 . Therefore, using the anti-MTH-Ag MAbs and anti-PSGL-1 antibodies, a comparative expression analysis of MTH-Ag and PSGL-1 was performed on BW- O and a range of BW-O-derived variants with differences in MTH-Ag expression and metastatic and organ-colonizing potential. BW-O-TUM, obtained after s.c. inoculation of BW-O (De Baetselier et al., 1988), and the T-cell hybridomas BW-19 and BW-O-LH (De Baetselier et al., 1984 b), all of which massively colonize spleen and liver after i.v. inoculation and exhibit high expression levels of MTH-Ag, reacted strongly with anti-PSGL-1 antibodies (Fig. 1 ). On the other hand, the MTH-Ag-negative variants BW-O, BW-N and BW-Sp3 did not show any positive signal with anti-PSGL-1 antibodies. Although BW-N and BW-Sp3 do tend to accumulate in the lymph nodes after i.v. inoculation and BW-Sp3 even forms infrequent metastases to the spleen, none of these last 3 cell-lines massively colonize spleen or liver. Hence, overall, PSGL-1 expression correlated with the MTH-Ag expression and the organ- colonizing potential of the various BW5147 variants.

Example 2: PSGL-1 transfected in MTH-Ag-negative BW5147 variants is recognized by anti-MTH-Ag antibodies

To definitely assess whether PSGL-1 is the antigen that is recognized by the anti-MTH-Ag antibodies, full-length PSGL-1 was amplified via PCR from the MTH-Ag-positive variant BW- 19, sequence verified, cloned in a mammalian expression vector and stably transfected into the MTH-Ag-negative variants BW-N and BW-Sp3. Upon screening of 50 neomycin-resistant clones of each cell-line, PSGL-1 surface expression was detected on 40-50% of the transfectants, of which 25-30% had PSGL-1 expression levels comparable to or higher than those on the metastatic BW-O-TUM variant. As shown in Fig. 2, the anti-MTH-Ag MAbs recognized PSGL-1 on the PSGL-1 transfectants and a perfect correlation was found between the signals obtained on the various clones with anti-PSGL-1 MAbs and anti-MTH-Ag antibodies, demonstrating that PSGL-1 is indeed the MTH-Ag.

Example 3: Expression of PSGL-1 in MTH-Ag-negative BW5147 variants does not result in enhanced malignancy

As a first approach to assess if the MTH-Ag/PSGL-1 is functionally involved in the metastatic behavior of BW5147 variants, we evaluated the metastatic properties after i.v. inoculation of several clones of the stable MTH-Ag/PSGL-1 transfectants of BW-N and BW-Sp3. Although the MTH-Ag/PSGL-1 expression levels on the tested transfectants were similar to or in some clones even higher than those on metastatic BW5147 variants such as BW-O-TUM (Fig. 2), no significant increase in organ-colonizing potential or reduction in survival time of mice was recorded for the MTH-Ag/PSGL-1 transfectants as compared to the parental cell-lines or empty vector-transfected mock clones (data not shown). Hence, if the MTH-Ag/PSGL-1 is functionally involved in the metastatic behavior of the BW5147 variants, it is clearly not sufficient on its own to confer spleen- and liver-colonizing potential to the BW5147 variants.

Example 4: RNAi-mediated down-regulation of MTH-Ag/PSGL-1 expression on MTH- Ag/PSGL-1 -positive BW5147 variants reduces their organ-colonizing potential As an alternative approach to assess the involvement of the MTH-Ag/PSGL-1 in the metastatic behavior of BW5147 variants, the MTH-Ag/PSGL-1 -positive BW5147 variants BW-O-TUM and BW-19 were stably transfected with 3 different PSGL-1 -specific siRNA constructs. For siRNA constructs 1 and 2, but not for construct 3, about 10% of the 50 neomycin-resistant clones of BW-O-TUM that were screened, exhibited a 30-50% reduction in MTH-Ag/PSGL-1 surface expression as compared to the parental cell-line. On one of the BW-O-TUM clones transfected with siRNA construct 2, MTH-Ag/PSGL-1 surface expression was reduced by more than 90% (Fig. 3). For BW-19, siRNA-transfected clones were obtained with a maximum of 40% reduction in MTH-Ag/PSGL-1 surface expression as compared to the parental BW-19. The effect of siRNA transfection on the MTH-Ag/PSGL-1 expression was fairly stable, since it was maintained even after more than one month of in vitro culture. As compared to the parental cell-lines, the siRNA-transfected clones did not exhibit significant differences in the expression levels of other membrane antigens (data not shown) that can affect the metastatic capacity of BW5147 variants, such as LFA-1 (Roossien et al., 1989) or

H-2D^ (VandenDriessche et al., A994). Moreover, the in vitro proliferation rates of the siRNA- transfected cells did not differ significantly from those of the parental cells, suggesting that siRNA expression had no overall toxic effects.

To assess the effect of MTH-Ag/PSGL-1 -down-regulation on the experimental metastatic capacity of BW-O-TUM, mice were inoculated i.v. with equal numbers of either i) the parental BW-O-TUM cells, N) a pool of 3 siRNA-transfected BW-O-TUM clones (one with construct 1 , two with construct 2) with, as compared to BW-O-TUM, less than 10% reduction in MTH- Ag/PSGL-1 expression (designated BW-TR10), iii) a pool of 2 siRNA-transfected BW-O-TUM clones (one with construct 1 and one with construct 2) with, as compared to BW-O-TUM, about 50% reduction in MTH-Ag/PSGL-1 expression (BW-TR50) or iv) the siRNA-transfected BW-O- TUM clone with, as compared to BW-O-TUM, more than 90% reduction in MTH-Ag/PSGL-1 expression (BW-TR90). As shown in Fig. 4A, the median survival time of the mice was 17-18 days post inoculation (d.p.i.), except for the BW-TR90-inoculated group, which exhibited a significant delay in mortality as compared to the BW-O-TUM-inoculated group (p<0.0001 ). At 17 d.p.i., moribund mice were sacrificed and the liver and spleen colonization was determined. As shown in Fig. 4B/C, less than 10% reduction in MTH-Ag/PSGL-1 surface expression (BW-TR10) had no significant effect on spleen or liver colonization. About 50% inhibition of MTH-Ag/PSGL-1 expression (BW-TR50) significantly reduced the liver colonizing potential, but had no effect on spleen colonization. At this time-point, also mice from an additional group inoculated with BW-TR90 were sacrificed. These mice did not exhibit any sign of spleen or liver colonization. At the time of mortality of the mice inoculated with BW-TR90 for the survival experiment, these mice did exhibit spleen and liver colonization, but in all mice the liver colonization remained very limited as compared to the levels observed with BW-O-TUM- inoculated mice sacrificed at 17 d.p.i. (Fig. 4B/C) and some of the BW-TR90-inoculated mice even did not form any metastatic nodules on the liver. Instead, most of these BW-TR90- inoculated mice developed a retroperitoneal tumor mass: in the back of the peritoneal cavity, behind the kidneys.

The results obtained with various siRNA-transfected BW-19 clones exhibiting about 40% reduction in MTH-Ag/PSGL-1 surface expression, were similar to those observed with BW- TR50: no delay in mortality and no reduction in spleen colonization as compared to the parental cells, but a significant reduction in liver colonization.

Collectively, the above data demonstrate that MTH-Ag/PSGL-1 plays a causative role in both liver and spleen colonization, with a more pronounced effect on liver metastasis.

REFERENCES

- Atarashi, K., Hirata, T., Matsumoto, M., Kanemitsu, N. & Miyasaka, M. (2005) J Immunol 174, 1424-32. - Ba, X., Chen, C, Gao, Y. & Zeng, X. (2005) J Cell Biochem 94, 365-73.

- Borges, E., Pendl, G., Eytner, R., Steegmaier, M., Zollner, O. & Vestweber, D. (1997) J Biol Chem 272, 28786-92.

- Chen, S. C, Huang, C. C, Chien, C. L, Jeng, C. J., Su, H. T., Chiang, E., Liu, M. R., Wu, C. H., Chang, C. N. & Lin, R. H. (2004) Blood 104, 3233-42. - Croce, K., Freedman, S.J., Furie, B.C. and Furie, B. (1998) Biochemistry 37 , 16472-16480.

- De Baetselier, P., Roos, E., Brys, L., Remels, L., Gobert, M., Dekegel, D., Segal, S. & Feldman, M. (1984a) Cancer Metastasis Rev3, 5-24.

- De Baetselier, P., Roos, E., Brys, L., Remels, L. & Feldman, M. (1984b) lnt J Cancer 34, 731 -8. - De Baetselier, P., Roos, E., van Hecke, D., Verschaeve, L., Brys, L. & Verschueren, H. (1988) lnt J Cancer 41 , 720-6.

- Fuhlbrigge, R. C, Kieffer, J. D., Armerding, D. & Kupper, T. S. (1997) Nature 389, 978-81 .

- Hayward, R. and Lefer, A.M. (1999) Shock 12, 201 -207.

- Hicks, A. E., Nolan, S. L., Ridger, V. C, Hellewell, P. G. & Norman, K. E. (2003) Blood 101 , 3249-56.

Hirata, T., Furukawa, Y., Yang, B. G., Hieshima, K., Fukuda, M., Kannagi, R., Yoshie, O. & Miyasaka, M. (2004) J Biol Chem 279, 51775-82.

- Katayama, Y., Hidalgo, A., Furie, B. C, Vestweber, D., Furie, B. & Frenette, P. S. (2003) Blood 102, 2060-7. - Kyriakides, C, Favuzza, J., Wang, Y., Austin, W. G. jr., Moore, F. D. jr. and Hechtman, H. B. (2001 ) Br J Surg 88, 825-830.

Laszik, Z., Jansen, P. J., Cummings, R. D., Tedder, T. F., McEver, R. P. & Moore, K. L. (1996) Blood 88, 3010-21 .

- McEver, R. P. & Cummings, R. D. (1997) J Clin Invest 100, 485-91 . - Moore, K. L, Stults, N. L., Diaz, S., Smith, D. F., Cummings, R. D., Varki, A. & McEver, R. P. (1992) J Cell Biol 1 18, 445-56.

- Moore, K. L., Patel, K. D., Bruehl, R. E., Li, F., Johnson, D. A., Lichenstein, H. S., Cummings, R. D., Bainton, D. F. & McEver, R. P. (1995) J Ce// β/o/ 128, 661 -71 .

- Roossien, F. F., de Rijk, D., Bikker, A. & Roos, E. (1989) J Cell Biol 108, 1979-85. - Sako, D., Chang, X. J., Barone, K. M., Vachino, G., White, H. M., Shaw, G., Veldman, G. M., Bean, K. M., Ahem, T. J., Furie, B. & et al. (1993) Cell 75, 1 179-86. Vachino, G., Chang, X. J., Veldman, G. M., Kumar, R., Sako, D., Fouser, L. A., Berndt, M. C. & dimming, D. A. (1995) J Biol Chem 270, 21966-74.

- VandenDriessche, T., Geldhof, A., Bakkus, M., Toussaint-Demylle, D., Brijs, L., Thielemans, K., Verschueren, H. & De Baetselier, P. (1994) Int J Cancer 58, 217-25. - Vanden Driessche, T., Verschueren, H. & De Baetselier, P. (1990) Invasion Metastasis 10, 65-85.

- Van Hecke, D., Verschueren, H. & De Baetselier, P. (1990) Int J Cancer 45, 773-82.

- Walcheck, B., Moore, K. L., McEver, R. P. & Kishimoto, T. K. (1996) J Clin Invest 98, 1081 - 7. - Wang, K., Zhou, X., Zhou, Z., Tarakji, K., Qin, J.X., Sitges, M., Shiota, T., Forudi, F., Schaub, R. G., Kumar, A., Penn, M.S., Topol, E.J., and Lincoff, A.M. (2002) Thromb Haemost 88, 149-154.

Claims

1 . The use of PSGL-1 to modulate metastasis of lymphoma.

2. The use of an inhibitor of PSGL-1 binding to repress metastasis of lymphoma.

3. The use of an inhibitor of PSGL-1 according to claim 2, whereby said inhibitor is an anti-PSGL-1 antibody.

4. The use of an inhibitor of PSGL-1 according to claim 2, whereby said inhibitor is mocarhagin.

5. The use of an inhibitor of PSGL-1 according to claim 2, whereby said inhibitor is soluble PSGL-1 .

6. The use of an inhibitor of PSGL-1 according to claim 2, whereby said inhibitor is soluble P-selectin.

7. The use according to any of the claims 2 - 6, whereby said metastasis is metastasis to the liver or the spleen.